Working paper NO.

WP 004/2013

Witches:
Through Changing Contexts Women Remain the
Target

Dev Nathan , Govind Kelkar and Shivani Satija

Institute for Human Development
New Delhi

New Delhi
2013
Published by:
INSTITUTE FOR HUMAN DEVELOPMENT
NIDM Building, IIPA Campus, IP Estate, New Delhi-110002
Phones: +91-11-2335 8166/ 2332 1610 • Fax: +91-11-2376 5410
E-mail: mail@ihdindia.org • Website: www.ihdindia.org

© Institute for Human Development, 2013

ISBN: 978-81-88315-40-6

Subscription Amount: ` 50/- /US $ 10

 Witches: Through Changing Contexts Women
Remain the Target

Dev Nathan,Govind Kelkar and Shivani Satija *

Introduction
This paper deals with the phenomenon of witch-hunting among indigenous peoples in
peninsular India. It looks at this phenomenon in a number of different contexts: the struggle
over domination in the sphere of rituals, and also contestation in more mundane spheres,
such as control over land and property and new forms of accumulation.
The phenomenon of witch-hunting is a form of violence, violence mainly against women
but also at times including other members of the family. As violence, is it transformative, an
attempt to change existing gender relations? Or, is violence constitutive of gender relations,
a phenomenon that exists as part of ongoing gender relations? We will argue that witch-
hunting has been both instrumental in changing gender relations, as well as constitutive of
gender relations, initially instrumental and then constitutive.
The contemporary condition of indigenous peoples is one of rapid change, a change
from subsistence economies to one of accumulation. The former is characterized by relative
stability and the latter by instability and growing inequality. We analyse the nature of witch-
hunts in the transitions that are underway.
What all this means is that there are various layers to the witch and witch-hunting
phenomena. But in order to set the scene for the analysis, we first summarize the dimensions
and extent of the witch phenomenon among Indian indigenous peoples. We concentrate on
the situation of indigenous peoples in Jharkhand, since that is the area we have studied more
extensively (see Kelkar and Nathan, 1991; Nathan 1997; Nathan, Kelkar and Yu Xiaogang,
1998).
Existence and Extent
Whatever disputes there may be about the existence of witch practices by women, the
denunciation and subsequent violence including killing, cannot be denied. This denunciation
of women as witches and their persecution in various forms, we will summarize under the
term ‘witch-hunting’. The prevalence of witch-hunting was much commented on by British
colonial administrators, such as Col. Dalton and W. G. Archer, and European missionaries,
such as S. Bodding in the 19th Century, in his Santhal Dictionary and J. B. Hoffman in the
20th Century in Encyclopaedia Mundarica.
There are partial figures in the immediate post-Independence period. For the district of
Malda, West Bengal, from 1950 to 1980 it was estimated that some 46 persons had been killed
as witches (Chaudhuri, 1987: 160). National police data have been recording ‘witchcraft’
under motives of murder since 2001.

* Respectively, Institute for Human Development, New Delhi, and Duke University, USA, Landesa, New Delhi, and
Institute for Human Development, New Delhi
2 IHD WORKING PAPER SERIES

Table 1 in the Annex is a summary of data from the National Crime Records Bureau
(NCRB). The NCRB data, unfortunately, are not gender disaggregated. But, we do know,
from various accounts that the victims killed are largely women, with men being involved
mainly as members of the household. There is some variation in this gendered picture among
the major indigenous communities of Jharkhand, with an almost entirely female incidence
among the Santhal, Ho and Munda, but less so among the Oraon people.
It should be noted that the data are likely to be an underestimate of the situation. Instances
of witch-killing could well be listed under other categories, such as property disputes or
personal vendettas. Nevertheless the numbers of witchcraft related killings are non-trivial.
The period from 2001 to 2012 averaged 168 witchcraft deaths1 nation-wide per year, with
a range from 114 deaths in 2004 to 242 deaths in 20112. The deaths were concentrated in
the states of Andhra Pradesh, Chhatisgarh, Jharkhand, Madhya Pradesh, Odisha and West
Bengal, with one or two also in Bihar, Gujarat and Haryana. As one would expect, they
were concentrated in the tribal-dominated states of peninsular India.
Jharkhand is home to the Association for Social and Human Awareness (Asha), an NGO
led by Poonam Toppo, whose grandmother was denounced as a witch. Its activities were
initially entirely focused on opposition to witch denunciation and witch-hunting, but have since
expanded to take up development issues of children’s education and ecological agriculture.
Asha has a figure of 371 witch killings in Jharkhand state from 2001 to 2008 (as reported
in the Hindustan, 2012). This is about 50 per cent more than the National Crime Records
Bureau (NCRB) figure of 250 killed in Jharkhand in the same period. Asha also estimates
about 100,000 women to have been charged with being witches, giving an average of a little
more than three witch accusations per village in the state’s 32,000 villages.
The official figures themselves are non-trivial and clearly warrant investigation and
explanation. The estimates of one person killed in every 100 villages and three accusations per
village are substantial. The phenomenon of witch-hunting cannot be explained by hysteresis,
or the continuation of a phenomenon that continues though its cause has ceased to exist.
There is a need to explain why it continues even in considerably changed socio-economic and
political circumstances, right into the current era of market-based globalization. Of course,
the Jharkhandi communities have not been immune to the impacts of market relations, with
a history dating back to the early colonial period of supplying timber and lac. In fact, it
was the destitution that followed a crash of lac prices that was the occasion for the Birsa
Munda-led revolt at the end of the nineteenth century.
Witches and Witchcraft
The first question that one has to deal with is that of existence of witches. Are there
really such people, such women? Talking to people of the region there certainly is a strong
belief in the existence of such women who practice a different form of worship labelled
witchcraft, and thus—as per popular belief— gain certain powers to cause harm. But are
there really such practices as gathering at night, in the sacred grove, where it is otherwise
forbidden for women to enter, usually on the night of a new moon (amavasya), dancing,
1
From the NCRB data we have taken both ‘murder’ and ‘culpable homicide’ to be witch killings.
2
For details see Table 1 in the Annex.
WITCHES: WOMEN AS TARGETS 3

possibly naked and uttering unknown chants? The Norwegian missionary, Bodding, who in
the later 18th century recorded what he could of Santhal life, wrote, ‘I am inclined to think
that the practice of witchcraft by Santhal women is, to a certain extent, really secret worship,
resorted to by women because they are not permitted to take part with the men directly and
personally in ordinary public worship,’ (1979: 224).
Some reputed contemporary anthropologists and observers have also mentioned in
personal communications that they have witnessed such forms of secret worship by women.
More recent accounts, though not first-hand, also testify to the continued existence of secret
forms of worship by Santhal and other indigenous women (see Bosu-Mullick, 2000 and
Mishra, 2003). What, however, has been noted as secret worship may well have been a
normal form of worship in an earlier period. In the course of history it may have moved
from centre-stage to the margins.
The Santhal and Munda both have myths of the manner in which the women-centred
religious practices were relegated to the margins or came to be denounced as negative.
Myths are not history, but they play a role in making a cultural account of how the history
was created. And, once mythologized, these notions continue to play a role in people’s
understanding of contemporary events.
The movement of religious practices to the margins is the result of a struggle, a struggle
in which formerly exalted practices and states, such as ‘being possessed’, are now derided
and even cast in a negative light. It is in this struggle that the women-centred practices and
rites came to be charged as being witchcraft. As Malcolm Gaskill points out, ‘… the charge
of witchcraft was never uniquely associated with one faith or another, but generated by
conflict between faiths,’ (2010: 16, emphasis in original). Discussed in more detail elsewhere
(Nathan, Kelkar and Xiaogang, 1998; and Bosu-Mullick, 2000) there was a conflict between
women-centred and men-centred religious practices among various indigenous peoples.
What is important for our present analysis is that these conflicts, mythologized women
as dangerous witches. There is an easy, even if unconscious polarization between ‘the
polarity of gender and the polarity of good and evil’ (Gaskill, 2010: 30). There surely is not
an invariant connection between women and evil, but many religio-cultural systems3 posit
such a connection among their founding metaphors.
Rebellion and Witch-hunting
Two periods of adivasi rebellion in Jharkhand were simultaneously periods of intense
witch-hunting. This was during the 1857 Mutiny, at which time the Santhal hul (rebellion)
led by the brothers Sighu and Kanhu took place and later at the end of the 19th century
when there was the Birsa Munda-led rebellion in the Khuti-Ranchi area. Preceding these
rebellions were periods of colonial rule. The colonial authorities, along with missionary
priests, obviously opposed witch-hunting and killing.
There was a belief that witches flourished under benign British rule. In Africa too
there was the belief that the colonial state protected witches by refusing to act against them

3
See Nel Noddings (1989) for an analysis of Eve and the origin of evil in Christianity.
4 IHD WORKING PAPER SERIES

(Geschiere, 1997: 15). In 1837, Wilkinson issued a directive with specific instructions against
witch murders (Sinha, 2007). When the political space of the colonial state shrank during
the 19th century adivasi rebellions, there was a rise in witch-killings. In a parallel manner,
one might note that the ending of apartheid in South Africa is said to have seen a rise in
witch-killings. ‘The problem seems to be that as soon as a new political space is opened
up, it is overrun by rumours about the use of sorcery and witchcraft’ (Geschiere, 1997: 7).
The Jharkhand adivasi rebellions combined anti-colonial and anti-women connotations.
As colonial rule weakened, many women were killed as witches. But should the anti-witch
drives of the adivasi rebels be seen as a type of James Scott’s ‘weapons of the weak’, as
argued by Shashank Sinha (2007)? It would rather be more straightforward to point out
that the adivasi rebellions of the 19th century also contained within them anti-women aspects
related to the traditional practice of witch persecution and killing.
Further, there was adivasi opposition to the witch-hunts (Ata Mallick, 2008). Even
in those days persecuted women and their families approached the colonial authorities for
help. Opposition to witch hunting then came from not just the colonial authorities alone,
but also from sections of the community itself. Since quite large numbers of women were
attacked and faced the threat of attack, there could have been substantial opposition. Chotrae
Deshmanjhi a participant in the Santhal rebellion of 1855 said, ‘We were all afraid seeing
such things. My two brothers suggested that we should leave the place immediately as we too
have women and girls. They might be identified as witches’ (ibid.: 118, emphasis added).
The Santhal myth, in particular, identified all women as actual or potential witches.4
By the time of the Adivasi Mahasabha in the 1930s and the Jharkhand Party in the 1950s
both Hinduism and Christianity had spread among the indigenous peoples in Jharkhand.
In addition, the attempt to create pan-Adivasi identities also led to ‘toning down internal
tensions (witch-killings included)’ (Sinha, 2012).
Nature of Witch -hunting
The types of events that constituted violence in witch-hunting are clear enough. They
involved humiliation (e.g. being forced to eat human or animal excreta), trashing of the
house, sexual and physical attacks on women and other members of the immediate family.
The witch and sometimes her family could be banished from the village. At times, the
women were killed.
Witch-hunting was not an inter-tribal matter; it did not involve members of one tribe
attacking those of another as having witches. In that sense, it was not similar to head-hunting,
which usually involved villages fighting each other. Witch-hunting was an internal matter,
an intra-tribe matter. Those making the accusations and those being accused belonged to
the same tribe. It was an internal struggle, a gendered struggle. Since any woman could be
denounced as a witch it would have served to terrorize and inhibit all women from being
forceful or insistent about anything.
More than just being members of the same tribe, the accuser and accused were usually
closely related to one another, members of the same kin group or lineage. The accusations
4
For an analysis of the Santhal myth of witch origin, see Kelkar and Nathan (1991); Nathan, Kelkar and Yu (1998)
WITCHES: WOMEN AS TARGETS 5

usually centred around losses suffered by one family, or section of the family. These losses
could be economic, such as of livestock or even the harvest; or the loss of family members
through illness. Any kind of misfortune was explained in terms of the malevolent intentions
of some persons, accused of using witchcraft to bring about misfortune.
These accusations were followed by the killing or driving out of the witch and the takeover
of her land and other property by the accusers, since, most often the accused women were
widows or otherwise single women. Such women had little familial support and thus could
be easy victims. A struggle within kin groups is often about property. Witch-hunting truly
is ‘the dark side of kinship’ (Geschiere, 1997: 11).5
In the struggle within kin groups, land was the central issue. In different tribes women
had varied rights to land as widows, ranging from the right to full-use, to the right to merely
be maintained. In the case of a widow without sons, the land then went to the nearest male
kin on the husband’s side. Those male kin with a claim over the land where often the ones
who initiated witch accusations in order to immediately takeover the land.
Land and Subsistence Economy
In the traditional tribal economy there was growth. A village could grow by bringing
new land under cultivation. A kin group could also grow by bringing more land under
cultivation. As sons got married and started new households they could bring new land
under cultivation. More land, which also meant more households, strengthened a kin group.
The economy was not stagnant—there was growth, but at the extensive margin. Growth
at the extensive margin meant that per capita income or consumption standards remained
more or less constant. Higher production which might occur in a good year did not mean
a higher consumption standard. The excess, over and above regular consumption or seed
needs, was often consumed in feasting. Or, at times the surplus produce would simply be
left to rot in the ground, as was observed in the case of swidden cultivators in Tripura
(Malabika Dasgupta, 2002).
When such an economy could grow by bringing more land under cultivation there may
have been less conflict over land within the village. But when colonialism set limits to the
expansion of cultivation, claiming the forests as state property, then there would have been
an intensification of land conflicts within the village. It should be noted that in both major
adivasi revolts in the 19th century, the right to expand cultivation into what were claimed
as state forests by the colonial state, was the key issue. The Chotangapur Tenancy Act that
resulted from the Birsa Munda-led revolt, explicitly recognized the right of villages over
their forests, legalizing the ‘khutkatti’ system.
The colonial state not only took over the forest lands stopping growth possibilities,
it also took steps against witch killing. As mentioned earlier, in 1853 Wilkinson (of the
famous Wilkinson Rules) criminalized witch killing, treating it as murder.
But even in a subsistence economy, there could be some possibilities for accumulation,
particularly through forms of trade, which colonialism made possible. For instance, in

5
This remark was made by Peter Geschiere (1997) about witchcraft in Cameroon or Africa in general.
6 IHD WORKING PAPER SERIES

Jharkhand there grew the trade in lac. In this subsistence economy, witch-hunting did
function as a levelling mechanism, in the manner that seems to have occurred in Africa
(Geschiere, 1997).
Among the Munda there was a system of redistribution that inhibited accumulation,
rather than centralized it, as was the case among the Khasi of Meghalaya.6 Any family that
accumulated wealth was subjected to hisinga or envy. The wealth was supposed to be due
to worship of the paltu or domesticated bonga or spirit. Success in accumulating wealth
was then due not to individual or household effort, but to the spirit, the bonga. At some
time the family would be required to give a feast to the whole village, which could leave
the family even worse-off than the rest of the village.
The village feasting was a manner of redistribution, or rather consumption of the
surplus. But it did not carry the negative connotation of wealth acquired through illicit or
negative means, as through witchcraft. Thus it was different from the African case where
any accumulation was due to witchcraft and led to persecution of the witch. Redistribution
by feasting was a relatively benign form of levelling, compared to denunciation as a witch.
Change in Production Orientation
Over time, the subsistence, or ‘full belly’-orientation of tribal production began to
change. New needs were created and grew, such as those for education and modern medical
services.With better communication the possibilities for trading opened up avenues for
increasing income.There was a shift from a subsistence form of production to accumulation.
However, the possibilities for accumulation were not equally shared by all households.
There were inequalities depending on access to external contacts, education, etc. Not
only were local goods, such as non-timber forest products (NTFP) traded, migration for
wage labour also opened up.
When there is growth of per capita income in an economy where growth is possible
only at the extensive margin, then that must come about through some forms of external
economic contact. Some persons got jobs outside the village economy, in some form of
migration. But with migration, and even more jobs, being limited, there could clearly be an
element of jealousy when some households, and not others, received increased income.
A shift from subsistence to an accumulative mode of production is a very major
transformation of the economic system. It is what may be called a civilizational change
(Nathan and Kelkar 2003). The change is, as mentioned above, not just in the types of
production carried out but also in the attitudes to production. Production is not just for
meeting relatively fixed needs, but now has the objective of maximizing income.
Such a shift in thinking about economic production also has many social consequences.
Reciprocity in sharing, works well when who does better is a random occurrence. Such
change and sharing tend to cancel each other out over time. But when some households

6
For an extended discussion of issues of accumulation among indigenous peoples see Nathan and Kelkar (2003).
WITCHES: WOMEN AS TARGETS 7

have a regular access to higher income than others in the kin group, then sharing tends
to break down. For instance, when one household has a member in a government job
or other such salaried job, then the claims of kin over that income become a source of
resentment. Those who do not have the benefit of having secured such a job become
envious of those who have.
Envy, or hisinga, as we saw above, is a well-established concept in tribal socio-economic
thinking. Earlier accusations of witchcraft did not include being economically better-
off as a factor. But more recent accounts, such as Archana Mishra (2003) and personal
communications with Poonam Toppo, of the anti-witch persecution NGO, Asha, mention
the rise of being better-off as a factor in witchcraft accusations.
Thus, there is a new kind of background to witch accusations. ‘The money she received
from him [migrant husband] every month became the root cause for her brother-in-law’s
jealousy,’ (Archana Mishra, 2003: 61); and in another instance, ‘We were leading a calm and
quiet life and my family was well-settled. Maybe this was the reason why they labelled me
as a witch,’ (ibid.: 95). These are new kinds of background elements in witch accusations,
of households that do well or better than others.
In these examples the witch could be thought to have used evil or at least occult
powers to secure some benefit for her family. The good fortune of these households is
directly linked to the absence of good fortune of other less advantaged households. Such
an analysis is explicitly made among the matrilineal Khasi of Meghalaya. When someone,
other than those from the clans of the chiefs (Syiems) or priests (Lyngdoh) accumulated
wealth, it was supposed to have been done by worshipping the serpent, u thlen. This
accumulation was tantamount to drinking the blood of other clan members.
The notion of sanctioned and non-sanctioned wealth is present among central Indian
tribes too. This is discussed in some detail in Bailey (1994), which we use over here to
supplement material from our earlier work.
There was the wealth from land. The village head and the village priest usually had
the better quality land in the village. If they were skilful and careful they could manage
some accumulation, though their feasting obligations would usually dissipate any possible
accumulation. But if some accumulation did take place, this was acceptable as a sanctioned
form of wealth. This wealth, one should note, was something entirely relative. It might just
mean having more grain, fruits and vegetables. Such extras could not really be accumulated
and were justified on the basis of the redistribution involved in various rituals.
The other way of accumulating wealth was a kind of ‘easy wealth’ (ibid.: 79). This was
due to luck, or more important, opportunistic behaviour or cheating. Even more important,
however, was the fact that such easy wealth was accumulated by those who were not
supposed to do so. This could be any family other than those of the headman and priest.
It could also extend to multi-ethnic villages, with a combination of tribes and castes.
8 IHD WORKING PAPER SERIES

Among7 the Munda and Ho, such a family is supposed to have tamed or kept the paltu
bonga, or domesticated spirit. In the multi-ethnic village that Bailey analyses, the family
is said to have kept a devata. Witches too have their own bongas. They could be used to
damage another person’s well-being. Between the witches and various bongas or devatas,
the point is that there were sanctions for acquiring wealth. If acquired it should be shared
with other members of the lineage. Those who did not do this voluntarily could be forced
to do this through the mechanisms of either hisinga (envy) or the witch accusation.
That acquiring wealth and not sharing it, could spark-off hisinga, is seen in one key
incident. The family of the sister-in-law of Ram Dayal Munda, one-time Vice Chancellor
of Ranchi University and a member of the Congress Party’s National Advisory Council
(NAC), was killed by a group of men from the village. Not only had they become better-
off, they had also given up drinking haria, or rice beer. Not sharing rice beer with your
brotherhood is even taken as an insult, a sign of anti-social behaviour and of looking down
upon other members of the brotherhood. They were asked to leave the village and when
they refused, all members of the family who were in the village were killed. There was no
witch accusation in this case.
If spirits could be used to accumulate wealth, they could also be used to damage
another person’s economic well-being. A market-based economy brings with it various
risks. Very damaging could be the loss of uninsured livestock; cattle in particular represent
large investments for agriculturist households.
With the growth of a new orientation towards accumulation away from subsistence
production, the witch could use powers to: (i) accumulate, and (ii) to damage others. Those
who did badly economically would try to find the witches who had damaged them. Those
who did well would be targeted as witches.
Witch-hunting, then, serves as a levelling tendency. Here the distinction between
witchcraft and witch-hunting is important. Witchcraft may be thought to be used for either
a positive reason, such as accumulation, or a negative reason such as to destroy another.
Overall though, witch-hunting is seen to serve only one tendency, that of levelling by not
allowing accumulation by some.
The above analysis is different from that by Peter Geschiere (1997) of witchcraft in
Cameroon. He thinks that representations of witchcraft strike a, ‘precarious balance between
“levelling” and “accumulating” tendencies’ (1997: 16). That is correct as far as witchcraft
goes, which can be used either to benefit oneself or harm others.
Witch hunting however would not have an accumulating tendency (unless it were
exclusively, or preponderantly used to target the worse-off sections and enable the
better-off to grab their land and other properties). A preliminary look at contemporary
cases of witch-hunting does not show that it is mainly the worse-off who are the victims

7
Thanks to Samar Bosu-Mullick for discussions on these points.
WITCHES: WOMEN AS TARGETS 9

of witch-hunting!
Further, as we saw about the Munda, the levelling function was served not only by
witch accusations but also by requiring those who had more to offer as a feast to the village.
Before going on to next section we would like to again reiterate that there is a difference
between two groups of those who get ahead of the rest. The first group of those who
enter into professions (or, even government jobs) and either completely move away from
the village or depend on it only in a nominal sense. There would certainly be a substantial
number of claimants on the income and property of such persons and families, but their
having moved away makes them somewhat immune to levelling pressures.
The other group is of those who have moved somewhat ahead but still remain
within the village economy. They may have some members who have migrated not into
professions but as workers at various levels. Their families would continue to reside in the
village and interact with them. This section is most vulnerable to levelling pressures. One
would expect that it is among them that women are likely to be denounced as witches
and subject to persecution.
Institutions and Economic Transition
It is important to look at the overall impacts of witch-hunting. The literature on witch-
hunting is trapped in an anthropological mode, if one may call it that. As a starting point it
is necessary to look at how those involved look at these institutions. But even when going
beyond how local people look at these institutions, it is necessary to place witch-hunting
within the context of the transitions actually underway in tribal economies. In summary
form, this is the transition from subsistence to an accumulating economy. In this transition
what is the impact of witch-hunting?
This impact can be broken down to a number of aspects in order to be analysed. At a
social level there is loss due to destruction of property, at least the witch’s house and all
materials in it. Land and other productive assets may not be destroyed in the course of
witch-hunting, but the house usually is trashed and made unfit for dwelling. In any case,
people are unlikely to stay in a declared witch-house.
Injury, banishment, or death, all result in a loss of labour power. There are seemingly
mundane matters like the cost of treatment and the loss of income due to the inability to
work during treatment and recovery, but these are real costs that must be attributed to
witch-hunting. Premature death, of course, is the ultimate loss of labour power, reducing
the productive potential of the household, lineage and tribe. With accumulation and the
possibilities of migration the loss of such labour power, whether temporarily through
injury or permanently through death, becomes particularly acute.
10 IHD WORKING PAPER SERIES

The above losses are concentrated in the households of the witches and through
that have an impact on production in the village or tribe. But more important may be the
impact of witch denunciation and witch hunting on the agency and initiative of women
and even the whole tribe. Any woman, particularly one who is successful or part of a
successful household, is in danger of being denounced and persecuted as a witch. This
would inhibit women’s’ agency in taking initiatives for economic betterment. It could also
lead to attempts to conceal actual accumulation. Such attempts could have costs at least
in terms of not investing in more visible, but higher-return arenas. Hiding money, rather
than investing it, has a cost.
The inhibiting effect on investment is not just a matter of women alone, but would
affect the households of which they are a part. Any successful household is likely to have
women denounced as witches. All households would then either restrict or hide their
accumulations and investments.
Accusations and witch-hunting would thus affect the agency of women and their
ability to use or even build upon their capabilities. These inhibiting effects spread from
women to their households, which in effect impacts the whole village and tribe. A resulting
low-level trap could be broken by a collective decision to take up new and promising
activities. But the continuation of suspicion and envy, hisinga, would make such collective
decisions difficult.
Finally, there is no doubt that witch-hunting and killing are a gross violation of human
rights. It may be argued that such practices are part of the culture of the concerned
indigenous peoples, but cultural practices have and do change.
Witch-hunting as Anti-modern
That witch-hunting continues to occur in a modern context does not, therefore, make
it a modern practice. We have seen that it is an anti-accumulative practice. That this is
framed in terms of levelling does not make it modern. Rather, one could argue that there
are two kinds of levelling. There is a levelling practice that inhibits accumulation and brings
down or levels those who get ahead of the rest. There can also be a levelling practice of
collective ownership based on accumulation in a market-based economy. Witch-hunting is
a levelling practice of the former type. Taking modernity to mean at least accumulation,
one can argue that witch-hunting is anti-modern at the economic level.
More important, however, is to frame modernity in terms of more equal gender
relations. On this count, there can be little doubt that witch-hunting and accusations are
thoroughly anti-modern. They inhibit the agency of women and are a threat to all and
any women of the tribe.
Witch-hunting in Changing Contexts
We have looked at witch-hunting in three different contexts. First is that of setting
up masculine domination in the ritual and political spheres. This is also the period of
displacement of women-centred religious practices and their replacement by men-centred
WITCHES: WOMEN AS TARGETS 11

religious practices. It is in this clash of religious and ritual systems that women’s’ religious
practices were denounced as witchcraft. The key aspect of this process was women not
having a voice in governance, in conjunction with their ritual devaluation.
The second context is that of the patrilineal descent system. Women’s land rights
were reduced over time. Witch accusations and the killing or driving out of witches were
among the forms of oppression used in this process, which established and re-established
women’s economic dependence.
Analytically we can distinguish these two spheres of political and economic affairs and
the context of political and economic domination by men and dependence of women.
But they may well occur simultaneously and at various points of time.
The third context of witch-hunting is that of transition from subsistence to an
accumulative economy, and the decline of reciprocity. In this context the target is not
just women, but may also include their households.
But across all three contexts the constant is the denunciation of women as witches and
their being hunted in various ways. This oppression of women in ritual, political, economic
and family or kinship spheres, or patriarchy, is the constant through all these contexts.
The violence of the witch-hunt seems to have had a constitutive role in establishing
women’s subordination. But it also has an instrumental role in continuing to keep women
subordinated. The likelihood of any woman being denounced as a witch has an effect on all
women, persuading them to follow established practices and not deviate from the norm.
Patriarchy strengthened over time, extending men’s domination from the ritual and
political spheres to the economic sphere. But with the growth of the market system and
the substantial possibilities for women’s independent economic activities we may see a
weakening of patriarchy. But this economic weakening needs to be accompanied by a
weakening of strongly held cultural beliefs of the need to control women.
Nature of Violence
Witch-hunting, even when it does not involve killing, is an extreme form of violence.
It can involve destruction of a house, being forced to eat human and animal excreta, and
physical and sexual assault. Either banishment or killing is often the end-result. With any
woman likely to be branded a witch, this violence is a threat against all women of the tribe.
Was witch-hunting and killing instrumental in converting gender difference into
gender hierarchy? Henrietta Moore (1994: 139) poses this question about violence about
women in general. Looking at the Santhal myth of the origins of women as witches and
men as witch-finders, one can certainly put forward the thesis that the violence of witch-
hunting and killing was instrumental in turning gender difference into gender hierarchy.
Thus this function of the violence of witch-hunting may be viewed as instrumental—that
is, violence employed to establish a patriarchal gender-hierarchy and its belief system.
12 IHD WORKING PAPER SERIES

But, having established gender hierarchy in the form of patriarchy, what then explains
the continuing violence of witch-hunts? One might use the term violence as constitutive
of gender relations to explain the continuing role of witch-hunts. This formulation is a
little different from Moore’s question: ‘does violence enter into the cultural construction of
personhood in different ways for women and men?’ (1994: 139). The violence of witch-hunting
clearly enters in different ways into the cultural construction of women and men in these
indigenous communities. Women are the victims (or, largely the victims), while men are the
perpetrators of that violence. Women are the witches, who can be subject to violence; while
men are the witch-finders who can initiate violence. Having played an instrumental role in
establishing gender hierarchy, the violence of witch-hunting then continues as a constitutive
part of gender hierarchy.
Dealing with Witch-hunting
Dealing with witch-hunting surely requires police action. But it is not necessary that
this will be a deterrent where people believe that their actions are justified by a moral
code or social norms that they adhere to. There have been reported instances where after
killing the woman, the perpetrators have turned themselves in at the police station. In
the case of the Munda family hisinga killing mentioned above, the killers went directly
to the police station. Nevertheless, police action alone is not sufficient. It needs to be
accompanied by a change in social norms.
How norms change is a difficult matter to analyse. In Weber’s analysis the Protestant
Ethic, a macro-level change in norms, brought about a change in norms and behaviour
at the micro-level. Saving and accumulation now become acceptable in social behaviour.
This is, in a sense, a top-down change—a macro-level change in beliefs bringing about a
micro-level change in norms and behaviour.
But there are also examples of changes in norms which have come about not through
top-down changes but more from a bottom-up movement. Micro-credit, when it started in
Bangladesh, went against the then established norm of women not being income-earners.
There were religious sanctions against women entering into micro-enterprises. As Jean
Ensminger and Jack Knight point out, ‘When we have sanctioning of deviant behaviour,
we have evidence for the existence of a norm,’ (1997: 3).
But over time, women’s participation in micro-enterprises has spread through the
country and a new norm of women contributing to household money earnings has come
to be accepted as a norm.8 “Some actors will eventually focus on a particular outcome and
others will in time follow suit, establishing a convention” (Ensminger and Knight, 1997:
3). With tens of millions of women running micro-enterprises, it amounts to that many
millions of women deviating from an older norm who in the process, ‘... make possible the
assertion of a new one’ (Keesing, 1975: 140, quoted in Ensminger and Knight, 1997: 1).
There is a social (and individual) selection of one norm over another. The change in
norms accompanies the actual change in women’s economic behaviour. The selection
of women’s micro-enterprises at both individual and social levels must mean that it is
WITCHES: WOMEN AS TARGETS 13

superior in providing a better economic result. In the economic context of poor women,
the fitness landscape would favour the selection of a norm that provides a better economic
outcome for poor women.
Ensminger and Knight put forward three mechanisms of changing social norms, ‘…
coordination of focal points, competitive selection among contracts, and bargaining’ (1997:
3). Focal points here would be those who take up activities resulting in accumulation. The
spread of these activities could be a factor in changing current norms. Whether those
activities spread or not, however, depends on whether they are likely to be selected by
large numbers of people. These activities should be economically superior to others in order
to be selected. These two mechanisms—of focal points and selection—then really turn
out to be just one mechanism, of being economically superior. An economically superior
practice is likely to spread, overcome old norms and, in the process, establish new norms.
Bargaining could also enter into the picture in the sense of undertaking some
redistribution through feasting to buy-off those who do not benefit from the new norms.
But there is a contradiction through such kin-based redistributive claims and incentives for
individual accumulation. The stronger the former claims, the weaker will be the incentive
to accumulate. More important than bargaining could be the possibility of drawing all or
most into the accumulative process. In the absence of such inclusion there are bound to
be tensions in accumulation. There will also be a contradiction between buying out those
not benefiting and incentives to accumulate. The substitution of state-based redistribution
through taxes in place of kin-based redistributions, rather than bargaining, could be a
way around resistance to the new norms.
The above analysis relates to dealing with the witch problem in so far as it relates to
the spread of a new economic system based on accumulation rather than a subsistence
economy. The inequalities resulting from individual property and capability-based growth,
exacerbate tensions in this transition. It is likely that a collective form of such a transition
would reduce the resulting tensions in such a transition. As analysed elsewhere (Nathan and
Kelkar, 1998 [Nathan, Kelkar and Yu]; and the Chonqing experience) collective economic
entities, even collections of entities owned by a village as a whole, are possible within a
market-oriented system of accumulation.
Moving beyond the strains of the civilizational transformation of economic systems,
there still remains the issue of gender relations. How does one deal with the gender-
relations issue, irrespective of whether it appears in the context of an economic transition,
a relatively stagnant land-based economy or an even more stagnant tea plantation labour
economy?
There have been some movements to oppose witch-hunting.
How do they frame (Goffman, 1974) the anti-witch hunt campaign?9 The category of
superstition or questioning the supposed existence of witches is one such frame. Others are

8
See Kelkar, Nathan and Jahan, 2003, and Nathan, 2005, for an analysis of this change in norms.
14 IHD WORKING PAPER SERIES

those of women’s development, whether through reducing their vulnerability or reducing

their dependence on men, and that of justice. The issue of development is not posed as
we have put it in this paper—as one of transition from subsistence to accumulation, but
the development schemes taken up (setting up women’s micro-enterprises) point in that
direction. In this regard there is NBPDC or North Bengal People’s Development Center is
working in the North Bengal tea plantations, while Jharkand NGO Asha has also expanded
its ambit from witch denunciation and witch-hunting, to take up development issues of
children’s education and ecological agriculture (Soma Chaudhuri, 2008).
Justice and the issue of superstition are related. They could be translated into the field
of culture. What is called superstition is the cultural belief in the existence of malevolent
women with extra-normal powers to cause harm. Linked to this is the attribution of
misfortune (e.g. death of a person or livestock) to acts by some human. There is the further
belief that unauthorized persons who accumulate wealth do so in an illegitimate manner,
a transgression compounded by their failure to redistribute their gains.
Thus the questioning of beliefs that support witch-hunting relate to some aspects of
the culture of indigenous peoples. It is often argued that such cultures and their practices
are not to be questioned, leave alone opposed. Nowadays in India, there is even a claim
to a right ‘not to be offended’ by criticism. But when cultural beliefs and their resulting
practices violate currently accepted notions of human rights, then it is human rights that
are superior to cultural practices.10

Future Research
In this paper we have put forward the thesis that the context of witch-hunting has
changed from the acquisition of land in a largely-subsistence, patrilineal economy to a
levelling tendency within systems of accumulation. It would be useful to conduct a survey
across a number of states to document the relative importance of these two contexts of
witch-hunting.

References
Appiah, Kwame Anthony, 2005, Identity and Ethics, Cambridge, MA: Harvard
University Press.
Bailey, F. G., 1994, A Witch-Hunt in an Indian Village: Or, the Triumph of Morality,
Delhi: Oxford University Press.

9
This section draws on Soma Chaudhuri (2008).
10
For a fuller discussion of this issue see Nathan et al. 2012. In a more general context, but one which is important
for framing the debate on human rights and culture see Appiah (2005).
WITCHES: WOMEN AS TARGETS 15

Bodding, Paul Olaf, 1979, ‘Some Remarks on the Position of Women among the Santals,’
in J. Troisi, ed, The Santals, Vol. 6, New Delhi: Indian Social Institute (mimeo).
Bosu-Mullick, Sanjay, 2000, ‘Gender Relations and Witches among the Indigenous
Communities of Jharkhand, India,’ in Gender, Technology and Development, 4, 3: 333-358.
Chaudhuri, A. B., 1987, The Santals, Religion and Rituals, New Delhi: Manohar
Publications.
Chaudhuri, Soma, 2008, Tempest in a Tea Pot: Analysis of Contemporary Witch Hunts
in the Tea Plantations of Bengal, PH.D. disseratation, Vanderbilt University, etd.library.
vanderbilt.edu/available/etd-07292008.../somawitchhunts.pdf, last accessed 3-2-2014.
Dasgupta, Malabika, 2002, “Objective Functions in Economic Models of Decisions
on Production; Evidence from Swiddeners in Tripura,”Economic and Political Weekly,
3 (340: 3556 – 9).
Ensminger, Jean and Jack Knight, 1997 ‘Changing Social Norms: Common Property,
Bridewealth and Clan Exogamy,’ in Current Anthropology, 38, 1: 1-24.
Gaskill, Malcolm, 2010, Witchcraft: A Very Short Introduction,Oxford: Oxford
University Press.
Geschiere, Peter, 1997, The Modernity of Witchcraft: Politics and the Occult in
Postcolonial Africa, Charottesville and London: University of Virginia Press.
Geschiere, Peter, 1998, ‘Globalization and Power of Indeterminate Meaning: Witchcraft
and Spirit Cults in Africa and East Asia,’ in Development and Change, 29: 811 – 837.
Holloway. Wendy, 1984, ‘Gender Difference and the Production of Subjectivity,’
in Henriques, J, Hollway, W., Venn, C., and Walkerdine, V. Changing the Subject,
London: Methuen.
Kelkar, Govind, 2000, ‘Witches: Asia,’ in Cheris Kramarae and Dale Spender, General
Editors, Routledge Encyclopedia of Women, Volume 4: 2040 – 2043.
Kelkar, Govind and Dev Nathan, 1991, Gender and Tribe: Women, Land and Forests
in Jharkhand, New Delhi: Kali for Women.
Kelkar, Govind, Dev Nathan and Rounaq Jahan, 2003, ‘Redefining Women’s ‘Samman’:
Microcredit and Gender Relations in Rural Bangladesh,” in Economic and Political Weekly,
August 7, 2004, 3627 – 3640.
Mallick, Ata, 2008, ‘Witch-hunting in 1857,’ in Economic and Political Weekly,
September 27: 118-19.
Miguel, Edward, 2005, ‘Poverty and Witch Killing,’ in The Review of Economic
Studies, 72, 4: 1153-1172.
Mishra, Archana, 2003, Casting the Evil Eye: Witch Trials in Tribal India, New
Delhi: Roli Books.
Moore, Henrietta, 1994, ‘The Problem of Explaining Violence in the Social Sciences,’
in Penelope Harvey and Peter Gow, editors, Sex and Violence: Issues in representation
16 IHD WORKING PAPER SERIES

and experience, London: Routledge, 138-155.

Moore, Henrietta and Todd Sanders,2001, editors, Magical Interpretation, Material
Realities, Modernity, Witchcraft and the Occult in Post-colonial Africa, London: Routledge.
Nathan, Dev, 1997, ‘The Creation of Patriarchy in the Transition from Tribe to Caste,’
in Dev Nathan, editor, From Tribe to Caste: History and Anthropology, Indian Institute
of Advanced Study, Shimla.
Nathan, Dev, 2005, ‘Transforming the Economic Ethic,’ in Economic and Political
Weekly, Jan. 27, 275-278.
Nathan, Dev and Govind Kelkar, 1997, ‘Collective Villages in the Chinese Market,’
in Economic and Political Weekly, May 3 and 10, 951-963 and 1037-1047.
Nathan, Dev; Govind Kelkar, 2003, ‘Civilizational Change: Markets and Privatisation
among Indigenous Peoples,’ in Economic and Political Weekly, May 17: 1955–1968.
Nathan, Dev, Govind Kelkar, Yu Xiaogang, 1998, ‘Women as Witches and Keepers of
Demons: Cross-Cultural Analysis of Struggles to Change Gender Relations,’ in Economic
and Political Weekly, October 31: WS-58 – WS-69.
Noddings, Nel, 1989, Women and Evil, Berkeley: The University of California Press.
Partners for Law in Development, 2012, Targeting of Women as Witches: Trends,
Prevalence and the Law in Northern, Western, Eastern and Northeastern Regions of India,
pldindia.org, accessed 20-7-2013.
Partners for Law in Development, 2013, Piecing Together Perspectives on Witch
Hunting: A Review of Literature, New Delhi: Partners for Law in Development.
Roy, Pamela, 1998, ‘Sanctioned Violence: Development and Persecution of Women
as Witches in South Bihar,’ in Development in Pratice, 8, 2: 136-147 Stewart, Pamela
H., Witchcraft, Sorcery, Rumors and Gossip.
Sinha, Shashank, 2007, ‘Witch-hunts, Adivasis and the Uprising in Chotanagpur,’ in
Economic and Political Weekly, May 12: 1672–1676.
Sinha, Shashank Shekhar, 2012, ‘On Witch-hunting’, The Sunday Guardian, November
3, 2012, sacw.net/article3299.html, accessed 20-9-2013.
Stewart, Pamela and Andrew Strathern, 2004, Witchcraft, Sorcery, Rumors and Gossip,
Cambridge: Cambridge University Press.
 Annex
Table 1: INCIDENCE OF KILLINF (MURDER AND CULPABLE HOMICIDE) DUE TO
WITCHCRAFT
States 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
ANDHRA 20 23 38 25 75 26 33 23 27 26 29 24
PRADESH
ASSAM 3 1 1 0 0 6 0 0 0 0 5 1
BIHAR 1 1 4 0 1 11 0 0 2 2 0 13
CHATTISGARH 16 4 14 11 9 10 8 15 6 8 17 8
WITCHES: WOMEN AS TARGETS

GUJARAT 2 4 1 10 6 3 1 0 4 5 0 3

HARYANA 2 0 0 0 28 34 30 26 32 58 5 0
JAMMU & KASH- 1 0 0 28 0 0 0 0 0 0 0 0
MIR
JHARKHAND 22 28 19 0 26 30 50 52 37 15 36 26
KARNATAKA 0 0 0 14 0 0 0 1 1 1 77 0
MADHYA 13 24 29 4 13 13 14 17 25 18 15 11
PRADESH
MAHARASHTRA 6 14 9 1 7 9 14 11 11 11 13 1
MEGHALAYA 0 0 0 22 1 0 0 3 1 2 3 0
ORISSA 30 39 26 0 25 36 28 23 36 34 41 32
PUNJAB 0 0 0 0 1 0 0 0 0 0 0 0
RAJASTHAN 0 0 1 0 0 1 0 0 0 2 1 0
SIKKIM 0 1 0 0 0 0 0 0 0 0 0 0
TAMIL NADU 0 0 0 0 0 1 0 0 3 0 0 0
TRIPURA 0 0 0 0 0 4 0 0 1 0 0 1
UTTAR PRADESH 0 0 0 0 0 3 0 1 0 0 0 0
WEST BENGAL 13 14 9 8 5 1 0 4 0 0 0 1
TOTAL STATES 129 153 151 114 197 188 179 176 186 182 242 121
17
 IHD WORKING PAPER SERIES
(New Series)
No. Authors Title
WP 04/2013 Dev Nathan, Govind Kelkar Witches: Through Changing Contexts Women
and Shivani Satija Remain the Target
WP 03/2013 Shivani Satija Violence Against Women in Rural Bihar- A case
of four villages
WP 02/2013 Sheila Bhalla Behind the Post 1991 ‘Challenge to The
Functional Efficiency of India’s Established
Statistical Institutions
WP 01/2013 Preet Rustagi, Dev Nathan, Women and Work in South Asia: Changes and
Amrita Datta and Ann Challenges
George
WP 05/2012 Amrita Datta, Gerry A Tale of Two Villages: Contrasts in
Rodgers, Janine Rodgers Development in Bihar
and B.K.N. Singh
WP 04/2012 Dev Nathan , Govind Kelkar Witches: Through Changing Contexts Women
and Shivani Satija Remain the Target
WP 04/2012 Janine Rodgers Labour Force Participation in Rural Bihar:
A Thirty-year Perspective based on
Village Surveys
WP 03/2012 K.P. Kannan How Inclusive is Inclusive Growth in India?
WP 02/2012 Sheila Bhalla Notes on Land, Long Run Food Security and
the Agrarian Crisis in India
WP 01/2012 Gerry Rodgers Understanding Unequal Economic and Social
Outcomes in Rural Bihar: The Importance of
Caste, Class and Landholding
WP 03/2011 Dev Nathan and Global Inequality, Rising Powers and
Sandip Sarkar Labour Standards
WP 02/2011 Preet Rustagi and Gender Asset Gaps and Land Rights in the
Rajini Menon Context of the Asia-Pacific Region
WP 01/2011 D. Narasimha Reddy NREGS and Indian Agriculture:
Opportunities and Challenges
WP 03/2010 Gerry Rodgers and Inclusion or Exclusion on the Periphery?
Janine Rodgers Rural Bihar in India’s Economic Growth
WP 02/2010 R Radhakrishna, C Ravi and Can We Really Measure Poverty and Identify
B Sambi Reddy Poor When Poverty Encompasses Multiple
Deprivations?
WP 01/2010 Alakh N. Sharma Political Economy of Poverty in Bihar: Nature,
Dimensions and Linkages
WP 03/2009 Dev Nathan and Markets, Technology and Agency: Indigenous
Govind Kelkar People and Change
WP 02/2009 Aseem Prakash Towards Understanding the Nature of Indian
State and the Role of Middle Class
WP 01/2009 C Upendranadh and Institutional and Governance Challenges
Rukmini Tankha in the Social Protection: Designing
Implementation Models for the Right to Work
Programme in India
WP 06/2008 Dipak Mazumdar and The Employment Problem in India and the
Sandip Sarkar Phenomenon of the Missing Middle
WP 05/2008 Aseem Prakash Social Conflict, Development and NGOs: An
Ethnographic Study
WP 04/2008 Balwant Singh Mehta and Wages and Productivity of Child Labour:
Kerren Sherry A Case Study of Zardosi Industry
WP 03/2008 Sheila Bhalla Scarce Land: The Cases of All India and West
Bengal
WP 02/2008 T.S. Papola and Market Access to Small Primary Producers:
R.P. Mamgain A Study of Vegetable Growers in the Supply
Chain
WP 01/2008 Preet Rustagi Rural Child Labour Markets in India: Nature of
Child Work Participation and Role of the Family
WP 03/2007 Alakh N Sharma Flexibility, Employment and Labour Market
Reforms in India
WP 02/2007 Preet Rustagi Rural Child Work, Labour and Daily Practices:
A Time Use Survey-based Analysis
WP 01/2007 R.P. Mamgain Growth, Poverty and Employment in
Uttarakhand
WP 04/2006 Sheila Bhalla Common Issues and Common Concerns in the
SAARC Region: Employment Generation and
Poverty Reduction
WP 03/2006 Rajendra P. Mamgain and Employment and Earnings in Uttaranchal:
Balwant Singh Mehta Trends and Policy Issues
WP 02/2006 Preet Rustagi Women and Poverty: Rural-Urban Dimensions
WP 01/2006 Dipak Mazumdar and Growth of Employment and Earnings in the
Sandip Sarkar Tertiary Sector, 1983-2000